Stack Overflow
Stack Overflow

Questions tagged [tfx]

TensorFlow Extended (TFX) is an end-to-end platform for deploying production ML pipelines

TFX is a Google-production-scale machine learning platform based on TensorFlow. It provides a configuration framework and shared libraries to integrate common components needed to define, launch, and monitor your machine learning system.

Resources

185 questions
2
votes
1 answer

How to reuse your TFX transforms for inference in production (streaming pipeline)?

I want to build a training pipeline using TFX, and eventually to reuse my data transformations to make inference requests to TensorFlow-Serving, which TFX is supposedly able to do. The TFX examples I found all seem to build a batch training pipeline…
Vincent Bernardi
  • 55
  • 4
2
votes
1 answer

tensorflow/serving with top n logits to return

I'm currently dealing with the challenge to serve my tensorflow models in a scalable way. As far as I know the recommended solution is to use the standard TensorFlow ModelServer. Common requirements are pretty well handled by this - but I want more.…
Tom
  • 33
  • 5
2
votes
0 answers

tfma.view.render_slicing_metrics not working

I am trying to visualize model results using tfma.view.render_slicing_metrics. I am getting the results with tfma.load_eval_result(tfma_artifact.uri) and I am able to print the results. [EvalResult(slicing_metrics=[((('trip_start_hour', 2),),…
Rathna Subramanian.T.A
  • 21
  • 2
2
votes
0 answers

How can I debug predictions on ML Engine, predictions returns empty array

I am implementing a tfx pipeline, similar to the chicago taxi example The prediction of the pushed model returns {"predictions": []}. How do I debug this issue? I can see logs of the predictions being made. But because it returns an empty array the…
Matthias Durivet
  • 21
  • 2
2
votes
2 answers

Transforming dates in tensorflow or tensorflow extended

I am working with Tensorflow Extended, preprocessing data and among this data are date values (e.g. values of the form 16-04-2019). I need to apply some preprocessing to this, like the difference between two dates and extracting the day, month and…
Martijncl
  • 105
  • 2
  • 10
1
vote
1 answer

What does tfma.metrics.MeanLabel do?

I am interested in what tfma.metrics.MeanLabe does, and how it should be used, and what the difference is between tfma.metrics.MeanLabe and tfma.metrics.MeanPredictio and tfma.metrics.MeanAttributions. I am not sure why there is no explanation about…
Alex
  • 1,194
  • 1
  • 12
  • 20
1
vote
0 answers

How to get the list of features along side their schema and stats using TFX

Let's say that I have the following, very straightforward pipeline: import os from tfx import v1 as tfx _dataset_folder = './tfrecords/train/*' _pipeline_data_folder = './pipeline_data' _serving_model_dir = os.path.join(_pipeline_data_folder,…
Mehran
  • 15,593
  • 27
  • 122
  • 221
1
vote
0 answers

How to get prediction labels with Tensorflow Serving Predict API?

I have a keras classification model saved remotely, and I use RESTful API to call the model on my site and make predictions. The JSON request is done as follows URL: POST https://my.site/models/v1:predict JSON: {"instances": [["...something useful…
sound wave
  • 3,191
  • 3
  • 11
  • 29
1
vote
1 answer

ml_metadata.errors.AlreadyExistsError: Given node already exists

I ran into a problem using TFX, MLMD, and Apache-Airflow as the orchestrator. Local-dag-runner, provided by TFX, works fine, resulting in distinct artifacts for each pipeline component run. The problem arises when airflow is used as the…
Parham Davari
  • 379
  • 4
  • 6
1
vote
0 answers

How to pass a *serialized* tensor to a TensorFlow Model Serving server?

I implemented a simple TF model. The model received a serialized tensor of a gray image (simply a 2d ndarray), and restored it to a 2-d tensor. After then, some inference is applied on this 2-d tensor. I deployed the mode with TensorFlow Model…
Zepp L.
  • 11
  • 1
1
vote
2 answers

TFX - What is example_gen_pb2 and where is it documented?

The ExampleGen TFX Pipeline Component has the example code which uses example_gen_pb2 but what is it and where is it documented? # Input has a single split 'input_dir/*'. # Output 2 splits: train:eval=3:1. output = proto.Output( …
mon
  • 18,789
  • 22
  • 112
  • 205
1
vote
1 answer

TFX - How to inspect records from CsvExampleGen

Question How to inspect the data loaded into TFX CsvExampleGen? CSV Top 3 rows from the california_housing_train.csv looks…
mon
  • 18,789
  • 22
  • 112
  • 205
1
vote
1 answer

Add reserved tokens to `tft.vocabulary`

I would like to append words to the vocabulary created by tft.vocabulary that are not a part of the training samples (i.e. and tokens). I see in the docs that the tft.vocabulary function can take an argument key_fn which the docs…
Zach Robertson
  • 11
  • 1
1
vote
0 answers

what is the v1 folder about in TFX

I'm new to TFX and a bit confused about what the v1 about, and what is its relationship with the code outside v1? Do I understand correctly that the official APIs are in v1, meaning I should always do import tfx.v1.something instead of just import…
zyxue
  • 7,904
  • 5
  • 48
  • 74
1
vote
0 answers

Customize Airflow / TFX Pipeline to upload Files in S3

I am currently trying to customize a tfx pipeline, which is being orchestrated by Airflow. Basically I want to upload all generated files by the DAG to a S3 bucket storage after the run of the pipeline. The issue is that every tfx pipeline component…
niklas_den
  • 19
  • 8
1 2 3
…
12 13